Add iterators to have more control over recognition process

[henrivain]

Result iterator

Tesseract result iterator gives more control over image result output.

Output

Result iterator gives access to different recognition block sizes that are

• Block
• Paragraph
• TextLine
• Word
• Symbol

Implementation order

• [x] Add dll imports
• [x] Implement Iterator logic with IEnumerator
• [x] Implement Iterable logic with IEnumerable
• [x] Implement higher level API like ITesseract
• [x] Add support to iOS
• [x] Add support to Windows
• [x] Add support to Android
• [x] Generate iOS dll import nuget
• [x] Generate new Nuget package version

Page iterator

Tesseract page iterator gives access to text location in image. This is secondary milestone.

• [x] Add dll imports
• [x] Implement Iterator logic with IEnumerator
• [x] Implement Iterable logic with IEnumerable
• [x] Add all platform support

[henrivain]

Development progress

Follow progress in add-recognizion-iterator-functionality -branch

Branch now runs TesseractOcrMaui/TesseractTestClass.cs -> RunAsync(); at startup for easier testing during development.

[henrivain]

Issue created from user request

Is it possible to get an array of lines rather than the whole text as a string ?

Tesseract returning hierarchy structure for OCR is (like Azure and friends do…):

Page
  Block
    Paragraph
      Line
         Word

The most efficient way would be a new function returning an array of Pages, 
each page would have an array of Blocks, each block an array of paragraphs and so on.

Doing this, we could have a real analyze of the content, to extract some identified values, 
and not only a “big string” where getting a word with its “meaning” is not really possible.

[henrivain]

Uploaded nuget for IOS dll imports https://www.nuget.org/packages/TesseractOcrMaui.IOS/1.1.0